A Simple Model of Physics and Consciousness

Prerequisites -

STOP - read the About page first.

Support -

This Simple Meditation arose and was instrumental in supporting the development of this paper.

Introduction

In this paper, we present a Simple Model of Physics and Consciousness. 

First, we present a simple model of physics. And then using the model, we explore simply, and step-by-step how consciousness can emerge from within it, giving rise to a simple model of physics and consciousness.

As with all models, this paper is less about accuracy in all details, but rather a simplification so that concepts can emerge with lucidity and clarity.

These models may have nothing to do with our physical reality or our consciousness. But our objective is to “start fresh”; and build a model of physics, and observe how consciousness can emerge within the model, and then, with that in mind, observe how well it aligns with our current understanding of our universe and our consciousness.

In the end, this paper strives to provide insight on a wide range of current topics, from simulation theory, to a functional roadmap to AGI and machine consciousness, to better clarity in understanding how our reality unfolds, and to clarity and comprehension of our own consciousness.

Certain things are out of scope for this paper, including the origin of the universe, the origin and ontology of mathematics, and so on. (Obviously.)

While the concepts here are simple, they can prove to be profoundly daunting psychologically and existentially. In fact, I believe the objections or difficulties with the models will have more to do with the existential implications than with the concepts themselves. Those are actually simple.

I hope that we can explore the concepts here as scientists with courage and clarity.


A Simple Model of Physics

Let’s start with the simple model of physics.

For this physics model, we want to start with a blank canvas. So it’s important to discard all thoughts of the laws of physics and nature that we know about in our actual universe. Let’s discard all concepts of matter, energy, space and time; all the laws of motion and gravity, all known laws of physics, and everything else. 

For our model building, we begin with a simple nothingness, a timeless and empty void. 

Then, within this nothingness, for simplicity, we begin with a set of two numbers. The specific two numbers are not important, so let's start with:

{ 7, 2 }

Imagine that is all there is. This is the entirety of our model universe, these two numbers. They are the beginning set of our simple model of physics. They are static, frozen, timeless and motionless. They may or may not represent anything. It doesn’t really matter “what” they are. They just “are”.

They are ALL and EVERYTHING that exist in our simple model universe.

We will call this set of numbers a “Universe Information Packet” (UIP).

The important point is that our UIP is a total and complete description of our universe in its entirety, down to the smallest detail. It is everything and there is nothing else. There is nothing in the universe that is not contained, described or modeled exactly via the UIP. You could say that whether the UIP is “real” or “representative” is a meaningless question because they are exactly the same and there is no way to tell the difference.

Also, whether the UIP contains “numbers” or “information” or “bits" or “reality” is also meaningless within the context of this model, since the description is exactly equivalent to the reality. A UIP can be arbitrarily large, and the information or data or “reality” within it can be encoded in many different ways, but the fundamental point is that the UIP contains a full and complete description of the universe in a frozen state, without energy, motion or time.

Imagine that you could press a “STOP” button in our universe so that everything instantly freezes into timelessness. Absolutely nothing happening, not even thought. Now imagine that you were somehow separate and could navigate the entire universe and record all the properties of every particle or waveform or entity of our universe in full and complete detail. You would record this full description of every atom and every particle in perfect, complete detail into your (very large) notebook.

That notebook transcription would be a UIP, a “Universe Information Packet” or “Universe State Description” for our universe.

(This is just a thought experiment. Our universe does not work exactly like this, certainly not at large scales. But I believe, and we’ll unpack further, that this is an extremely precise model for exactly what evolution sees and saw as biological life unfolded).

The Universe Information Packet (UIP) is the first component of our simple model of physics.

Next, we introduce the second component of our simple model of physics, the Universe Rule Engine (URE).

The Universe Rule Engine performs a “simple” task.  It takes a UIP as input, applies its rules to it, and outputs a new UIP. 

The original input UIP is discarded, and the new output UIP becomes the “current”, “real”, or “live” UIP. Then the new/current UIP is fed into the URE, and it is discarded as the URE outputs another new, current UIP.  

This process continues over and over again, indefinitely; infinitely. But only one UIP is ever “real”, “live”, “current” or “actual”.

It can be thought to occur quickly, or it can be thought to take a very long time for a URE to ingest a UIP and calculate and output the new one. From the perspective of “inside” the UIP it does not matter, since it is frozen and timeless. The “actual” time needed to compute the next UIP is a meaningless question from within the UIP.

Going back to our first simple UIP, the numbers 7, 2 could represent anything (or nothing). We could imagine that the 7 is a point on a number line, and the 2 represents a “stride”. 

So our example UIP { 7, 2 } is fed into the URE and a new UIP {9, 2} emerges and {7, 2} is gone. {9, 2} is fed in and {11, 2} emerges, and then {13, 2}, {15, 2}, {17, 2}, {19, 2} and so on, ad infinitum. 

For a slightly larger example, consider these three UIPs. Each contains three number pairs, and the URE again simply adds the second to the first, like so:

{ 7, 2           {9, 2 {11, 2

  3, 4       >   7, 4. > 11, 4

  2, 1 }          3, 1} 4, 1}

Of course, a UIP can be arbitrarily large - up to and beyond the size of our known universe. 

Maybe the URE has a rule that when a number reaches a certain point the sign is reversed so that it bounces back and forth. The rules of the URE could be simple or complex.

Again, imagine that the universe had a STOP button. You press STOP and our entire universe freezes into a timeless, frozen, static state. You record all the properties, the position, angle, and momentum of all the particles and all of the energy waves and photons, electrons and quarks in the universe. This full description creates a UIP. 

You take this UIP and feed it into our universe’s URE. This URE contains and executes our universe’s laws of nature and physics. So our laws of nature are applied to the UIP, and out pops a new UIP. This new UIP becomes the current UIP as the old one is discarded. Then the new UIP is fed in, and another new one pops out, and the process repeats, over and over again, forming our experienced universe.

As we consider this model, some salient points emerge, both about our universe and about this modeled universe.

  • The first is the unmistakable sense of an eternal or perpetual now. It is an apparent fact that our universe is in a perpetual “now” state between what just was, and what is just going to be. We are in an “eternal now”. There is only now. In some sense, the past and future are not “real” in the way that “now” is real.

  • This actual now moment, the only thing that actually exists, is extremely “thin”.

  • There are an enormous number of “UIP transitions” compressed into what we call time. A million a second? Sure, probably. Ask your local physicist.

  • Within this model, time can be considered not “real” but instead an “emergent” phenomenon of a transition between states.

  • This modeled universe is inherently quantized. At the lowest level, there is a “jump” between states rather than a continuous uninterrupted flow from one state to another - like a second hand on a clock that “ticks” rather than “sweeps”. Of course these jumps would be far too small for us to detect in our normal lives.

  • This modeled universe is Markovian. That is, the URE considers only the ONE prior UIP. If the URE is ingesting UIP #53, that is the ONLY thing being considered. UIP #1 to #52 are not considered in the construction of #54. They don’t exist.

  • So this modeled universe is also forgetful and lossy. What is gone is gone. Although certainly, depending on the rules of the URE, certain states could persist over many UIPs. Similarly there is not some grand plan, some grand long term intention. There is only an “unfolding”.

So here we have proposed two elements -  a Universe Information Packet (UIP) that describes the frozen, static state of the universe, and a Universe Rule Engine (URE) that takes a UIP as input, applies its rules to it, annihilates it, and produces a new UIP, forming a continual “eternal now” universe.

Together, these two elements form our Simple Model of Physics.

Conceptually simple, existentially HARD. To understand the rest of the paper, these must be internalized.

A Simple Model of Consciousness

Basic Brain Biology in a UIP/URE universe

To think about consciousness in a UIP/URE universe, let’s start by think about brain biology. Nature and evolution designed and architected the brain as exactly the type of device or apparatus that is needed in a UIP/URE paradigm. In basic brain biology we have synapses and neurons. Neurons are the “buckets” and synapses are the “pipes” between the buckets, forming many-to-many connections between the neurons.

Imagine that the URE has two rules:
#1 - What’s in the synapse flows to the neuron.
#2 - If the neuron reaches 1 it discharges out to 0.

So an extremely simplified UIP/URE view of basic neurobiology would look something like this:

UIP #405
Incoming Synapse #1: 0.2
Incoming Synapse #2: 0.2
Neuron: 0.7
Outgoing Synapse #1: 0.0
Outgoing Synapse #2: 0.0

UIP #406
Incoming Synapse #1: 0.0
Incoming Synapse #2: 0.0
Neuron: 1.1
Outgoing Synapse #1: 0.0
Outgoing Synapse #2: 0.0

UIP #407
Incoming Synapse #1: 0.0
Incoming Synapse #2: 0.0
Neuron: 0.0
Outgoing Synapse #1: 5.5
Outgoing Synapse #2: 5.5

This is an extremely simplified view but it paints the picture well. Brain biology has coopted the URE and “knows” its rules to implement its functionality. (Of course, in the actual URE these rules are written in the language closer to atoms and chemistry, not neurobiology.)

Entities in UIP/URE Universes

Now let’s consider entities in a UIP/URE universe. In a UIP/URE universe, an “entity” can be best described as a “collection of elements” in one UIP that “wishes” to persist together in future UIPs. The precise definition of what exactly is and is not part of an entity, from a UIP perspective may be ambiguous or hard or impossible to precisely define. Looking for a precise definition of identity within UIP structures is essentially an incorrect paradigm. UIPs are state descriptions and UREs process them. Entities may have meaning to themselves, but not to the UIP/URE itself. (An entity may be a “Markov Blanket”.)

Bootstrapping

Designing the first/original brain in a UIP/URE is a very difficult challenge, but the good news is that apparently nature, biology and evolution have already done this for us, so we are going to ignore bootstrapping for now, for simplicity, although admittedly this feels a bit like cheating. We are going to “jump forward” to where we have an “entity” in a UIP/URE world, and lo and behold, this entity already has a brain apparatus. Hooray, how convenient!

The Five Functions

In basic brain functionality in entities in UIP/URE universes, there are 5 processing functions. They are Observation, Memory, Modeling/Prediction (with a Learning sub-function), Goal/Preference selection, and Action Potential. Each of these can be well defined in UIP/URE information processing and we’ll take each of them in turn.


Definition

For simplicity we will now define and use the term “Walle” (from the pixar movie). A “Walle” is an entity in a UIP/URE universe that already contains a brain apparatus implemented within it. A Walle is a singular UIP that we can reason about. And we can use the term “Walle” to refer to the entity, and the brain within the entity interchangeably.

Function #1 - Observation

For Walle, observation means: Salient features of PRIOR UIPs that now exist within Walle. Or the flow of salient features from UIPs flowing into Walle. In humans of course, observation is mediated mostly by sight - photons, eyes, retinas, ocular nerves, and sound - air vibrations striking eardrums and cochlea. Several “profound” features of observation appear:

  • These observations are always about PRIOR/past UIPs - we’re always behind!

  • Everything is internal. We think we “see the world” but it’s “all in our minds”.

  • Obviously observation will always be a small, reduced portion of the UIP. There will always be much much more information in the UIP than can possibly fit in Walle. This also raises an attention and awareness problem. Which observations are meaningful and relevant?

  • There is obvious subconscious information processing occurring. Sound and music and speech come to us as a stream of tonal vibrations. But our conscious awareness doesn’t hear this. We hear “music” or “a story”; a higher level processed information stream.
    For example, if we look at a made up sentence: Glixon tendo smard - our brain needs to see each letter, convert them into phonetic sounds, and put them together into a word/sentence stream. But we “get” it essentially instantly. Did you actually notice the “r” in the word “smard” before now? No? Then it was subconscious.

So, beyond the sidebars, Observation for Walle simply means: some information about prior UIPs is now in Walle.

Function #2 - Memory

For Walle, memory means that these observations persist UIP to UIP. If an observation brings information into Walle, but these are lost in subsequent UIPs, it is “forgotten”. But if the information persists, it is “remembered”.

Function #3 - Modeling and Prediction

For Walle, now that he has Observation and Memory, he can start Modeling and Predicting. That is, his brain system can start to say: Ok, I have these memories, and these observations, so therefore I predict that future observations will have salient features such and such.

This immediately brings in the sub-feature; Learning. The brain system predicts that a future UIP will have such and such a feature. It DOES, YES! Strengthen that model. It does NOT, NO reduce or rework that model.

Function #4 - Goal/Preference Selection

For Walle, now that he has Modeling and Prediction capability, he also has the ability to articulate goals and outcomes and express his preferences for them. That is, he can list out numerous models and predictions, assign preference and likelihood scores to them and then shuffle the “best” preference to the top.

Function #5 - Action Potential

Now that Walle has a selected best preference at the top of the list, he can trigger an action potential. This is mediated by the same basic brain biology as all the rest, except that instead of feeding to other neurons, these feed to nerves and muscles (or for Walle - servos and circuits) that mediate and facilitate the desired action outcome.

Functional Purposes

  • The purpose of observation is to create memory (so that salient features of the current UIP persist in future UIPs because the current UIP state will be gone.)

  • The purpose of memory is to create models and predictions of what future/upcoming UIPs might look like

  • The purpose of the models and predictions is to sort them through a preference filter/matrix.

  • The purpose of sorting things through a preference filter/matrix is to organize them and find the best preference.

  • The purpose of finding the highest preference one is to activate action potential in that direction.

  • The purpose of activating action potential is so that the entire entity’s future state in subsequent UIPs is different than what it would have been had no action taken place.

  • The ultimate purpose of taking these actions is so that the entity continues to persist in future UIPs.

A few notes - 

  • Evolution and bioinformatic systems in brains, mediated via URE rules as the UIPs transition create this.

  • This is 100% deterministic.

  • In miniature, basic implementations, this is NOT conscious. This is simple robotics. This is an insect.

Consciousness Emerging

In a miniature, basic implementation, this is NOT conscious.

But let’s watch what happens as some of the five functions become more and more and more robust -

Observation - essentially no changes required

Memory - imagine memory becoming extremely robust, able to remember across many time domains (seconds, minutes, hours, days, weeks, years) and across many different knowledge domains; people, tasks, world navigation, etc. with extreme richness and depth and clarity. Obviously this would require a much bigger brain for Walle.

Modeling and Prediction - imagine this becoming richly sophisticated, able to accurately model and predict across many domains and across many timescales, with models that are deeply self-aware and also deeply aware of “others as essentially similar to self”.

Preference Selection - if Walle has excellent observation and rich memory and sophisticated and precise modeling and prediction, then he should be exceptionally well equipped to make excellent choices toward goals. (Exactly what those goals should be is a little out of scope for this specific paper, but of extraordinary importance to think about).

And finally, action potential. And here too there is essentially little that needs to be said. Rather it would be a discussion more about Walle’s physical body and capabilities that execute his action potential with excellence.

Candidly, as I think about these five functions growing and emerging into richness, I see something that is clearly more conscious and more capable than we humans ourselves. It seems to me that at some point in development we will say “clearly Walle is conscious”. A conundrum is that we can never truly say, “YES, this other entity truly conscious”. We can only affirmatively say with 100% confidence “I am conscious”. But if we are “strict” about this we then must also say, “I’m not sure that my dear friend is conscious” and I don’t believe any of us would say that.

So consciousness, it seems to me, exists on a continuum. Certainly “insect Walle” doesn’t seem to be conscious.

We lose consciousness as we sleep. We’ll use terms like “drifted out of consciousness”. Sleep, it seems, is a requirement for biological brain health and maintenance. And interestingly, dreams are “forms of consciousness” that seem to occur when our brains are undergoing the maintenance. And as we wake in the morning, we may be groggy and incoherent until slowly we become “sharp” by mid-morning. Unfortunate diseases like Alzheimer's seem to slowly rob us of robust and sharp consciousness.

A salient feature of consciousness seems to emerge around the benchmark of: can model oneself, and others as similar to oneself in the world. Or: includes a sense of self and others in their world models. A newborn baby cannot do this, and I think most reasonable people would agree that, though exquisitely precious, a newborn baby is not conscious. And I also think most reasonable people would agree that by around the age of seven (and earlier) certainly all healthy bio-typical children are able to do this and are of course fully conscious.

A simple functional example

Here is a simple functional example from our lived reality for demonstration purposes. Imagine we are walking down the street and we see a child pick up a rock and throw it at us. We quickly duck and the rock sails over our head. After we shake off our warranted anger at the child, we can see all five elements clearly. Let’s model them as UIPs.

First, we observe the child throwing the rock. But that fact will quickly fade into history, into subsequent UIPs, so it’s essential that we remember this as the rock is in flight. It’s also important to be able to model the rock’s flight path to determine if it will impact us in future UIPs. Then we have to model if we prefer to be hit by rocks or not. That’s a NO. So we have to take action and duck so the rock flies past us. Whew.

In this simple example, we find all five elements: observation, memory, modeling, preference and action. Notice how quickly and instinctively all five features appear. In our universe, this ballet of the rock in flight and our cascading biological mechanism to avoid it unfold instantly but over millions of UIPs.

Being Conscious

So what is it like to be conscious? It’s the HUM of UIPs and the HUM of observation, memory, modeling, preference and action. As we take a bite of chocolate, the observations of our taste buds flow that sweet taste into our brains. Chemical pathways light up and burst into neurobiological hums. We remember many of the prior chocolates we’ve had and how this compares, perhaps somewhat subconsciously. We are also processing vision perceptions and observations, forming memories and retrieving memories, making plans, thinking about our relation to others, what we want tomorrow and realizing we need to go to the bathroom. That’s consciousness.

Simple Model of Consciousness

So our simple model of consciousness is biological brains, in a UIP/URE universe observing, remembering, modeling, planning, choosing preferences and taking actions in rich and robust ways, fully aware of ourselves in our models, and also aware of “others similar to self”.

And this completes our Simple Model of Physics and Consciousness. What a journey this has been. The concepts are simple but the existential work of fully integrating them is profoundly hard.

Wishing you all the best on your journey.

Love and care,

BWM